2. Results and Discussion
2.5. Electronic and Steric Properties of MICs and their precursors
The detail understanding of the electronic and steric properties of ligands is essential to estimate their ability for catalytic applications. One of the most established methods for the determination of the electron donor ability of ligands is the Tolman Electronic Parameter (TEP) method (chapter 1.5). Experimental and calculated values are known for a variety of ligand systems. Herein, the reported values (entry 1–9) are compared with the calculated values of the synthesised compounds (entry 10–14).
Table 2-10: Calculated and experimentally determined TEPs for selected NHCs and those presented in this work. Experimental values of entry 1-9 for the complexes (NHC)IrCl(CO)2 and (NHC)Ni(CO)3 were reported in a review article by Nolan et al.[124] DFT
calculated values for entry 1-6 have been reported by Gusev.[158] The values have a maximum deviation of 2.5 cm-1. Entry NHC (NHC)IrCl(CO)2[cm-1] (NHC)Ni(CO)3[cm-1] DFT calculation[cm-1]
1 SIPr 2051.1 2052.2 2051.5
2 SIMes 2050.8 2051.5 2051.5
3 ItBu 2048.9 n/a 2050.6
4 IMes 2049.6 2050.7 2050.6
5 IPr 2050.2 2051.5 2050.5
6 IAd 2048.3 n/a 2045.8
7 IPr-2,4-Ph 2038.4 n/a n/a
8 IPr-2-Ph-4-(Ph-4-OMe) 2039.2 n/a n/a
9 IPr-2-Ph-4-(Ph-4-Me) 2038.3 n/a n/a
10 (IPrPh-2-Me)(13) n/a n/a 2051.9* 2137.4#
11 (IPrPh-4-Me)(12) n/a n/a 2052.6* 2138.4#
12 (IPr-4-OMe)(14) n/a n/a 2051.2* 2141.6#
13 (IPrPh-4-CO2Me)(15) n/a n/a 2052.7* 2141.3#
14 (IPrPh)(7) n/a n/a 2040.1**
2052.2* 2138.7#
*Values for entry 10–14 are calculated with the BP86/def2-SVP functional.
#Values for entry 10–14 are calculated with the B3LYP/def2-SVP functional. The standard deviation for both calculations is around 1 cm-1.
**Value for entry 14 was calculated following the instructions by Gusev to establish comparability.
In Table 2-10 some representative NHCs (entry 1–6), MICs (entry 7–9) and the synthesised compounds from this work (entry 10–14) are shown. DFT calculations were carried out according to Gusev, which allows extracting a comparison of the TEP values. Gusev calculated the TEP for 76 NHCs with the same DFT method and supported his results with experimental data.
Furthermore, experimentally determined IR values are presented in order to classify the recent results.[158] Imidazolidinylidenes show larger TEP values than imidazolylidene and are therefore
Results and Discussion
at the C2- and C4-position (entry 7–9) leads to increased electron density in the zwitterionic imidazole moiety. The more substituents with a positive inductive effect are bound to the imidazole moiety the more electron rich are the compounds which leads to a lower TEP value.
The prepared ligand is an effective electron donor but due to the two backbone H-atoms maybe not stable enough to isolate the free MIC. The calculated TEP values for entry 10–13 have to be handled with care. The first value was calculated with the BP86 DFT-functional and the second was calculated with B3LYP. Between these calculations are great differences, which could be seen comparing the values 2051.85 (BP86) and 2137.4 cm-1 (B3LYP) for entry 10. As not enough compounds have been compared using those calculation strategies, these compounds could not be classified concerning their donor ability. But a comparison between entries 10–13 could be done and within the standard deviation of 1 cm-1 the substituent at the phenyl ring seems to make no significant difference for the donor strength. Comparing the backbone CH-shift of C2- and C4- substituted NHCs a significant difference can be observed. It is likely to see that a substitution at the C4-position has more influence on the acidity of the C5 H-atom. The backbone protons of (IPrPh)I and IPr·HCl show a 1H-NMR resonance in the range of 8.0–8.3 ppm. The C2-proton of IPr·HCl has a resonance at 10.1 ppm, which is in much lower field and therefore much more acidic than the backbone protons. Furthermore, the adjacent hydrogen atom and the carbene could switch among each other. This is not possible for the nNHC due to the adjacent threefold coordinated nitrogen atoms.
Coincidentally Dipp Migration
Attempts for the preparation of the nickel tricarbonyl and the iridium dicarbonyl chloride NHC compounds were made to experimentally investigate the CO stretching vibration (i.e., TEPs).
Unfortunately, these efforts were not successful. The reaction of Ni(COD)2 with the MIC could just be performed in situ due to the instability of the free MIC.
N
Figure 2.27: Reaction of 7a with KHMDS and nickel cyclooctadiene led to a Dipp N-C migration.
As known from earlier reports a migration of the C2-bound substituent to the C4-backbone position is likely to happen.[43h, 119, 138a]
Surprisingly, a migration of the Dipp substituent of the nitrogen atom to the C4-backbone position has been so far not observed.
Results and Discussion
Figure 2.28: Migration of Dipp ligand from the nitrogen to the backbone of the imidazole moiety. The data of this crystal structure is below our internal quality standards and is therefore not further discussed but should illustrate the unexpected result.
The synthesis of the iridium complex also did not lead to the desired compound. The [Ir(COD)Cl]
dimer could be synthesised according to literature.[31] Unfortunately, the [(NHC)Ir(CO)2Cl]
complex could not be prepared due to the instability of the free carbene.
2.5.2. NMR Measurements
On comparing the 1H- and 13C-NMR chemical shift values of the five different C2-arylated imidazolium salts, it becomes obvious that the substitution of the initial aryl halide exert an influence on the backbone H-atoms of the imidazole moiety (Table 2-11). As discussed for compounds 7–10 the C4/C5 backbone shift can be correlated with the CH acidity. The further their downfield shift the higher the acidity. This gives us the possibility to synthesise new C2-arylated compounds with other substituents and tune the acidity as needed.
Table 2-11: 1H- and 13C- NMR shifts of the backbone hydrogen- and carbon- atoms respectively of compounds 7, 12–15.
1H-NMR shifts of the backbone H-atoms [ppm]
13C-NMR shifts of the backbone C-atoms [ppm]
(IPrPh-4-OMe)I(14) 8.01 126.52
(IPrPh-4-Me)I (12) 8.02 126.71
(IPrPh-2-Me)I(13) 8.11 127.16
(IPrPh)I(7) 8.15 127.09
(IPrPh-4-CO2Me)I(15) 8.30 127.85
Compound 15 is therefore the most acidic compound compared to the other synthesised compounds.
2.5.3. Percent Buried Volume Calculation
To investigate the ligand environment further occupation calculations with the program SambVca
[32]
Results and Discussion
by Cavallo et al. were used for the calculation.[32] Not surprisingly, the values for both cases show the same tendency. The substituent has just a very little influence on the %Volbur. Four out of five aryl halides are para substituted and have a minor impact on the strain of the Dipp ligand, which occupies the carbene sphere. The ortho substituted aryl halide shows the highest %Volbur which can be explained by the steric effect of the ortho methyl group on the Dipp ligand. One Dipp ligand is not involved in the occupation of the coordination sphere which generally leads to smaller %Volbur values compared to nNHCs. Due to the expanded N–C–N angle one Dipp ligand is pushed into the coordination sphere.
Table 2-12: %Volbur values for selected compound and a schematic representation of the buried volume..
N
R M d N
Compound %Vbur %Vburopt
(IPrPh-2-Me)I 35.6 33.4
(IPrPh)I 33.9 31.7
(IPrPh-4-Me)I 32.3 31.4
(IPrPh-4-CO2Me)I 32.1 31.2
(IPrPh-4-OMe)I 31.1 30.3
[(IPrPh)Cu]I 30.3 -
[(IPrPh)Cu]I, 1.897 33.3 -
IPr 33.6 -
In terms of sterics, C2-arylated compounds do not differ much from a nNHCs. Comparing IPr with the (IPrPh)I compound, the %Volbur value just differs by 0.3% which is negligible. The different reactivity of compound 7a towards deprotonation and the problems of isolating the MIC IPrPh allow for the conclusion that the acidity of the backbone protons plays a major role in terms of reactivity.
Results and Discussion